root-project
GPU runner for CI
Today, I heard that are GPUs runner in the CERN Gitlab are available: https://gitlab.cern.ch/rbritoda/gitlab-runner-public Technically, it possible to use this runners for GitHub. We do it in our projects, e.g. cupla: https://github.com/alpaka-group/cupla/tree/dev
It works over Gitlab.com. The GPU runners can be registered on GitLab.com and they are available for GitLab-CI jobs. To make the runner available on GitHub, you need to mirror the GitHub repo to Gitlab. Each time, if there is a change on GitHub, it will automatically copied to GitLab.com, there it triggers the CI and the result is send back to GitHub.com.
@Axel-Naumann @vgvassilev @reikdas Could this a possibility for the Cling-CUDA mode? Maybe it is also possible to use the GPUs over the CERN Jenkins.
Hi @SimeonEhrig I am the creator of Cirun.io, "GPU" and "CI" caught my eye.
FWIW I'll share my two cents. I created a service for problems like these, which is basically running custom machines (including GPUs) in GitHub Actions: https://cirun.io/
It is used in multiple open source projects needing GPU support like the following:
https://github.com/pystatgen/sgkit/ https://github.com/qutip/qutip-cupy
It is fairly simple to setup, all you need is a cloud account (AWS or GCP) and a simple yaml file describing what kind of machines you need and Cirun will spin up ephemeral machines on your cloud for GitHub Actions to run. It's native to GitHub ecosystem, which mean you can see logs/trigger in the Github's interface itself, just like any Github Action run.
Also, note that Cirun is free for Open source projects. (You only pay to your cloud provider for machine usage)
Hi @SimeonEhrig I am the creator of Cirun.io, "GPU" and "CI" caught my eye.
FWIW I'll share my two cents. I created a service for problems like these, which is basically running custom machines (including GPUs) in GitHub Actions: https://cirun.io/
It is used in multiple open source projects needing GPU support like the following:
https://github.com/pystatgen/sgkit/ https://github.com/qutip/qutip-cupy
It is fairly simple to setup, all you need is a cloud account (AWS or GCP) and a simple yaml file describing what kind of machines you need and Cirun will spin up ephemeral machines on your cloud for GitHub Actions to run. It's native to GitHub ecosystem, which mean you can see logs/trigger in the Github's interface itself, just like any Github Action run.
Also, note that Cirun is free for Open source projects. (You only pay to your cloud provider for machine usage)
Hi @aktech,
Thanks for the information. It looks quite interesting. But I'm not sure if AWS and GCP provides a free content, which is enough for a CI and if not I'm not able to decide it, because I'm not a member of the CERN. But really interesting work.